ai.canshu.com
中文
VRAM Calculator Advisor & TCO API Pricing Compare Models Leaderboard GPUs

Search models, GPUs, and tools

↑↓ Navigate · Enter Open · ★ Watchlist prioritized ai.canshu.com
64 GB VRAM 1200 GB/s Bandwidth 376 TFLOPS FP16 · 310W

华为昇腾 Ascend 910B 64GB (HBM2e) Local LLM Benchmark & Guide

Runs 16 open-weight models cleanly · Ref $13,200 (~$206/GB VRAM)

Top Runnable Models on 昇腾 910B 64G by Use Case

Click to load ready-to-run CLI config
Best for Coding & Agents ~14.5 tok/s
Qwen3.5-27B

Coding score 81 · fp16 (57.72/61.4 GB)

Open Launch CLI →
Best for General Chat & Reasoning ~255 tok/s
Qwen3.5-35B-A3B (MoE)

Arena Elo ~1458 · q8_0 (39.63/61.4 GB)

Open Launch CLI →
Best for Vision & OCR ~255 tok/s
Qwen3.5-35B-A3B (MoE)

Supports Text+Image+Video · q8_0 (39.63/61.4 GB)

Open Launch CLI →
Concurrency (Batch):
MoE 35B / A3B·40L (d=2048)·16Q/2KV·Ctx 256K·Text+Image+Video
Runnable Models →
GPUs
64 GB VRAM·1200 GB/s·Safe Usable 61.4 GB
Smooth Fit23.34 / 61.4 GB~420 tok/s
3. Weight Quantization
4. Per-Request Context16K (16,384 tok)
Verified: 2026-11-20·Roofline bandwidth/compute estimate (Batch=1)
Smooth Fit
23.34/ 61.4 GB usable

Physical 64 GB · Load 38%

Theoretical Decode Speed
~420tokens/s

MoE: reads only 1.82 GB active weights/step

Max Context at Batch=1
256K/ 256K

Enable Q8/Q4 KV quantization to scale batch concurrency

Weights: 21.22 GBKV Cache (1×16K): 0.4 GBOverhead: 1.72 GB
🚀 Blazing Fast (Instant CoT)~1.2s for a 500-token response
Reference Launch CLI:
OLLAMA_CONTEXT_LENGTH=16384 ollama run qwen3.5:35b-a3b

昇腾 910B 64G Fits Qwen3.5-35B-A3B (MoE) · Cost & TCO

Approx. $13,200 (~$206/GB VRAM) · 310W TDP

Checking current market prices for 华为昇腾 Ascend 910B 64GB (HBM2e)? Search Price on Amazon →

📋 All Open-Weight Models Tested on 昇腾 910B 64G (Q4_K_M, 16K)

Model Q4 VRAM Fit Status Est. Speed Best Quant Report
Qwen3.5-35B-A3B (MoE) 23.34 GB Smooth Fit ~420 tok/s q8_0 View →
Qwen3.5-27B 19.33 GB Smooth Fit ~50.7 tok/s fp16 View →
Gemma-4-31B-IT 22.47 GB Smooth Fit ~43.9 tok/s q8_0 View →
Gemma-4-26B-A4B-IT (MoE) 17.92 GB Smooth Fit ~320.3 tok/s fp16 View →
Qwen3-32B 25.03 GB Smooth Fit ~39.2 tok/s q8_0 View →
DeepSeek-R1-Distill-Qwen-32B 25.03 GB Smooth Fit ~39.2 tok/s q8_0 View →
GLM-4-32B-0414 21.26 GB Smooth Fit ~43.4 tok/s q8_0 View →
Qwen3-30B-A3B (MoE) 20.92 GB Smooth Fit ~307.9 tok/s q8_0 View →
Llama-3.3-70B-Instruct 49.48 GB Smooth Fit ~19 tok/s q5_k_m View →
Qwen3-14B 12.29 GB Smooth Fit ~83.8 tok/s fp16 View →
DeepSeek-R1-Distill-Qwen-14B 12.82 GB Smooth Fit ~81.6 tok/s fp16 View →
Qwen3-8B 7.94 GB Smooth Fit ~139.8 tok/s fp16 View →
Gemma-4-E4B-IT (Edge Multimodal) 6.25 GB Smooth Fit ~289.8 tok/s fp16 View →
DeepSeek-R1-Distill-Qwen-7B 6.09 GB Smooth Fit ~170.2 tok/s fp16 View →
Qwen3.6-35B-A3B (MoE) 23.96 GB Smooth Fit ~420 tok/s q8_0 View →
Qwen3.8-27B 19.83 GB Smooth Fit ~49.3 tok/s fp16 View →
Qwen3.5-122B-A10B (MoE) 77.26 GB Partial RAM Offload (Slow) ~40.2 tok/s q2_k View →
Llama-4-Scout-109B-A17B (MoE) 71.92 GB Partial RAM Offload (Slow) ~26.8 tok/s q3_k_m View →
Kimi K3 (2.8T MoE) 1733.2 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
DeepSeek-V4-Pro (1.6T MoE) 990.02 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
GLM-5 (744B-A40B MoE) 461.99 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
Kimi K2.6 (1T MoE) 620.83 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
Qwen3.5-397B-A17B (MoE) 247.5 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
DeepSeek-R1 (671B-A37B MoE) 416.53 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
GLM-5.3-Flash (320B-A18B MoE) 199.39 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
Qwen3-235B-A22B (MoE) 148.98 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
DeepSeek-V4-Flash (284B-A13B MoE) 176.19 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
MiniMax-M3.1-Flash (428B MoE) 267.98 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
Step-3.5-Flash (196B MoE) 123.02 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
Kimi K3.1 Pro (2.8T MoE) 1722.14 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
Qwen3.8-2.4T-A95B (MoE) 1515.15 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
Meituan LongCat-2.5 (1.78T MoE) 1101.87 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
Meituan LongCat-2.0 (1.78T MoE) 1099.4 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
Kimi K2.7 Code (1T MoE) 636.9 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
Kimi K2.5 (1T MoE) 636.9 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
Xiaomi MiMo-V2.6-Pro (1T MoE) 641.56 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
Mistral Large 4 (1T-A52B) 621.03 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
GLM-5.3 (753B MoE) 467.56 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
GLM-5.2 (753B MoE) 467.56 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
Step 5 Preview (600B MoE) 376.18 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
MiniMax-M3 (427B MoE) 267.36 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
DeepSeek-V4.1-Flash (1017) 236.39 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
Xiaomi MiMo-V2.6-Flash (311B MoE) 196.03 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
ERNIE-5.5-Flash (280B MoE) 178.12 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
MiniMax-M2.7 (229B MoE) 146.22 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
Step 3.7 Flash (201B MoE) 126.47 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →
Qwen3.8-Flash-Next (180B MoE) 113.11 GB Out of VRAM (OOM) — Cloud / Multi-GPU View →

Frequently Asked Questions

What is the largest LLM 华为昇腾 Ascend 910B 64GB (HBM2e) can run smoothly?

With 64 GB VRAM, 昇腾 910B 64G comfortably runs 4-bit (Q4_K_M) models up to Qwen3.5-35B-A3B (MoE) (35B parameters).

How fast is 昇腾 910B 64G for LLM token generation?

Driven by its 1200 GB/s memory bandwidth, 昇腾 910B 64G generates roughly 420 tokens/s on Qwen3.5-35B-A3B (MoE) (Q4_K_M).